Why did my Back-UPS Pro die this way?

[robdejonge]

Twelve years ago I bought an APC Back-UPS Pro 900 (BR900GI) and although it has survived for a decent amount of time, it has not been smooth sailing all along. 

In 2016, after 4 years of use, the battery died. This didn't just happen, but somehow at the same time as the battery dying, one of the connected hard drives also died. Which caused which I don't know, but I'm inclined to blame the UPS as power is its business. Data was permanently lost as a result and I had to buy a new hard drive; thanks APC! I've replaced the batteries every 3-4 years since. 

October of 2023 and again Feb of 2024, the unit started screaming and reported an F05 error. In both cases, all connected equipment got inelegantly disconnected from power. Can't be good for those devices. I've been told the 'charge module' needs replacing, and doing so is prohibitively expensive given I'm due for a battery replacement too. 

But it raised the question .... if the charge module needs replacing, why did the UPS disconnect my devices? Surely it can just report an error but keep passing on grid power to my devices? I mean ... "you had one job!" kind of situation in my book. 

Not a typical end user question, but I'm curious to learn more!

[Teken · Accepted solution]

Short Answer: Everything is tied together in the main circuit.

Long Answer: There are units that incorporate a manual bypass. This allows the unit to receive AC Mains directly bypassing a large portion of its internal circuitry.

UPS’s that incorporate a manual bypass are normally seen in the larger (expensive) enterprise units. 

NOTE: Regardless of (IF) a unit has a manual bypass this should not be counted on for back up or redundancy. ☝️ The only sure way is to have two UPS’s connected to a (ATS) Automatic Transfer Switch. Should the primary unit fail the ATS will transfer over to the secondary UPS for power.

Keeping in mind a ATS is NOT going to help you at all if AC Mains is still present but there is an active fault / issue with the primary unit! 

The only methods to alleviate such an issue is either (physically) monitoring the unit for a live fault and manually engaging the ATS. Or to use SNMP Trap information sent to a NMS (Network Monitoring System) to help track, monitor, notify, and react.

The NMS would send a command to the ATS to switch over to the secondary UPS to provide backup power. 

At the end of the day it makes sense to upgrade to a new unit vs repair. If you do it’s worth considering a unit that provides a pure sine wave AC output vs modified (step) sine wave. Don’t worry too much about units that offer so called PFC (Power Factor Correction) circuitry. I would focus more on units that allow a user to adjust for a wider (low / high) voltage transfer, surge protection (higher joules) with a lower clamping voltage of 330 VAC vs a unit that clamps at 550 and above! 👎

Consider a unit that also offers Ethernet surge protection to help protect your network. Ensure the RJ45 protection is for gigabit speed (1000) and not just for 10/100.

More expensive units incorporate better RFI / EMI filtering to provide cleaner power to the connected load. 

AVR: You should really understand how the Automatic Voltage Regulation (AVR) works and operates in any UPS you’re considering. As it relates to APC they use a combination of *Sensitivity* that can be set to (Low, Med, High) and upper / lower Voltage Transfer settings to tell the unit to engage the battery only circuit.

What does that mean?!?

It means it impacts the service life of the battery cells and how often you need to replace the same! It’s a balance of providing clean power vs battery service life.